A new quadrature sampling technique for arbitrary bandpass signal within baseband sampling rate is presented. The input bandpass signal whose carrier frequency lies in the A/D baseband sampling rate is first decimated...A new quadrature sampling technique for arbitrary bandpass signal within baseband sampling rate is presented. The input bandpass signal whose carrier frequency lies in the A/D baseband sampling rate is first decimated by factor 2 and modulated by (- 1)n, and then is interpolated by a linear phase FIR all-pass filter, finally the modulated complex envelope of bandpass signal can be produced.展开更多
In order to search for the seismic wave characteristics of low frequency signals in the Alxa Left Banner region,Inner Mongolia,the low frequency signals of seismic wave data are extracted from the earthquakes of MS5. ...In order to search for the seismic wave characteristics of low frequency signals in the Alxa Left Banner region,Inner Mongolia,the low frequency signals of seismic wave data are extracted from the earthquakes of MS5. 8 in 2015 and MS5. 0 in 2016 in this area. The results show that:① Before the MS5. 8 earthquake,the seismic stations located near the epicenter in Wuhai,Dongshengmiao,and Shizuishan recorded seismic waves that showed the phenomenon of spectrum shift from high to low frequency.② The low frequency signals recorded by different stations have obvious difference.③ According to the data recorded by the station closest to the epicenter,low-frequency signals were recorded about120 hours before the earthquake and had obvious anomalies. This may reflect slow slip before the earthquake.展开更多
传统的数据包络分析(Data Envelopment Analysis,DEA)模型无法比较多个有效决策单元(Decision Making Units,DMU)之间的效率高低,建立超效率DEA模型,可更好地对2011年江苏省10个城市承接离岸服务外包的投入产出效率进行实证分析。结果表...传统的数据包络分析(Data Envelopment Analysis,DEA)模型无法比较多个有效决策单元(Decision Making Units,DMU)之间的效率高低,建立超效率DEA模型,可更好地对2011年江苏省10个城市承接离岸服务外包的投入产出效率进行实证分析。结果表明:无锡、南京、苏州、镇江、泰州、常州综合效率值有效,但差距显著;南通、徐州、连云港和宿迁综合效率值无效,且效率偏低;无锡、南京、苏州和镇江处于规模效率递减状态,其他城市规模效率为递增状态,但均存在投入要素比例不合理问题,因而优化投入产出结构是提升离岸服务外包效率的关键。展开更多
针对低信噪比(signal to noise ratio,SNR)低截获概率(low probability of intercept,LPI)雷达脉内波形识别准确率低的问题,提出一种基于时频分析、压缩激励(squeeze excitation,SE)和ResNeXt网络的雷达辐射源信号识别方法。首先通过Cho...针对低信噪比(signal to noise ratio,SNR)低截获概率(low probability of intercept,LPI)雷达脉内波形识别准确率低的问题,提出一种基于时频分析、压缩激励(squeeze excitation,SE)和ResNeXt网络的雷达辐射源信号识别方法。首先通过Choi-Williams分布(Choi-Williams distribution,CWD)获得雷达时域信号的二维时频图像(time-frequency image,TFI);然后进行TFI预处理降低噪声干扰和频率维的位置分布差异,以适应深度学习网络输入;最后在ResNeXt基础上加入扩张卷积和SE结构提取TFI特征,实现雷达辐射源分类。实验结果表明,SNR低至-8 dB时,该方法对12类常见LPI雷达波形的整体识别准确率依然能达到98.08%。展开更多
The vibration signal contains a wealth of sensitive information which reflects the running status of the equipment. It is one of the most important steps for precise diagnosis to decompose the signal and extracts the ...The vibration signal contains a wealth of sensitive information which reflects the running status of the equipment. It is one of the most important steps for precise diagnosis to decompose the signal and extracts the effective information properly. The traditional classical adaptive signal decomposition method, such as EMD, exists the problems of mode mixing, low decomposition accuracy etc. Aiming at those problems, EAED（extreme average envelope decomposition） method is presented based on EMD. EAED method has three advantages. Firstly, it is completed through midpoint envelopment method rather than using maximum and minimum envelopment respectively as used in EMD. Therefore, the average variability of the signal can be described accurately. Secondly, in order to reduce the envelope errors during the signal decomposition, replacing two envelopes with one envelope strategy is presented. Thirdly, the similar triangle principle is utilized to calculate the time of extreme average points accurately. Thus, the influence of sampling frequency on the calculation results can be significantly reduced. Experimental results show that EAED could separate out single frequency components from a complex signal gradually. EAED could not only isolate three kinds of typical bearing fault characteristic of vibration frequency components but also has fewer decomposition layers. EAED replaces quadratic enveloping to an envelope which ensuring to isolate the fault characteristic frequency under the condition of less decomposition layers. Therefore, the precision of signal decomposition is improved.展开更多
文摘A new quadrature sampling technique for arbitrary bandpass signal within baseband sampling rate is presented. The input bandpass signal whose carrier frequency lies in the A/D baseband sampling rate is first decimated by factor 2 and modulated by (- 1)n, and then is interpolated by a linear phase FIR all-pass filter, finally the modulated complex envelope of bandpass signal can be produced.
基金the Major Scientific andTechnical Project of Department of Science and Technology,Inner Mongolia in 2016(Strong Earthquake Track in the Short Stage and Integration Innovation of Stereoscopic Observation Technology in Space and Ground)
文摘In order to search for the seismic wave characteristics of low frequency signals in the Alxa Left Banner region,Inner Mongolia,the low frequency signals of seismic wave data are extracted from the earthquakes of MS5. 8 in 2015 and MS5. 0 in 2016 in this area. The results show that:① Before the MS5. 8 earthquake,the seismic stations located near the epicenter in Wuhai,Dongshengmiao,and Shizuishan recorded seismic waves that showed the phenomenon of spectrum shift from high to low frequency.② The low frequency signals recorded by different stations have obvious difference.③ According to the data recorded by the station closest to the epicenter,low-frequency signals were recorded about120 hours before the earthquake and had obvious anomalies. This may reflect slow slip before the earthquake.
文摘传统的数据包络分析(Data Envelopment Analysis,DEA)模型无法比较多个有效决策单元(Decision Making Units,DMU)之间的效率高低,建立超效率DEA模型,可更好地对2011年江苏省10个城市承接离岸服务外包的投入产出效率进行实证分析。结果表明:无锡、南京、苏州、镇江、泰州、常州综合效率值有效,但差距显著;南通、徐州、连云港和宿迁综合效率值无效,且效率偏低;无锡、南京、苏州和镇江处于规模效率递减状态,其他城市规模效率为递增状态,但均存在投入要素比例不合理问题,因而优化投入产出结构是提升离岸服务外包效率的关键。
文摘针对低信噪比(signal to noise ratio,SNR)低截获概率(low probability of intercept,LPI)雷达脉内波形识别准确率低的问题,提出一种基于时频分析、压缩激励(squeeze excitation,SE)和ResNeXt网络的雷达辐射源信号识别方法。首先通过Choi-Williams分布(Choi-Williams distribution,CWD)获得雷达时域信号的二维时频图像(time-frequency image,TFI);然后进行TFI预处理降低噪声干扰和频率维的位置分布差异,以适应深度学习网络输入;最后在ResNeXt基础上加入扩张卷积和SE结构提取TFI特征,实现雷达辐射源分类。实验结果表明,SNR低至-8 dB时,该方法对12类常见LPI雷达波形的整体识别准确率依然能达到98.08%。
基金Supported by National Natural Science Foundation of China(Grant Nos.51175316,51575331)
文摘The vibration signal contains a wealth of sensitive information which reflects the running status of the equipment. It is one of the most important steps for precise diagnosis to decompose the signal and extracts the effective information properly. The traditional classical adaptive signal decomposition method, such as EMD, exists the problems of mode mixing, low decomposition accuracy etc. Aiming at those problems, EAED（extreme average envelope decomposition） method is presented based on EMD. EAED method has three advantages. Firstly, it is completed through midpoint envelopment method rather than using maximum and minimum envelopment respectively as used in EMD. Therefore, the average variability of the signal can be described accurately. Secondly, in order to reduce the envelope errors during the signal decomposition, replacing two envelopes with one envelope strategy is presented. Thirdly, the similar triangle principle is utilized to calculate the time of extreme average points accurately. Thus, the influence of sampling frequency on the calculation results can be significantly reduced. Experimental results show that EAED could separate out single frequency components from a complex signal gradually. EAED could not only isolate three kinds of typical bearing fault characteristic of vibration frequency components but also has fewer decomposition layers. EAED replaces quadratic enveloping to an envelope which ensuring to isolate the fault characteristic frequency under the condition of less decomposition layers. Therefore, the precision of signal decomposition is improved.
